| In this dissertation a general equilibrium approach is developed to model the determination of efficient monitoring. The same underlying theoretical approach is applied to two distinct cases where there is missing information about the true characteristics of agents or their actions. Both essays explore the public good aspects of monitoring for control or identification of agents and consider externalities imposed on one type of agent by another because differences between agents are not readily observable.;The first essay, "Monitoring, Productivity and Worker Utility: Determining Efficient Monitoring," develops a model of the labor market where workers differ in their inherent productivity and tolerance for being monitored, but this difference is not directly observable by firms. Monitoring increases output, but has a negative impact on all workers' utility. The model establishes the conditions for optimal monitoring and evaluates the relative stability of pooling and separating equilibria. The model provides an explanation for a number of labor market phenomena, including inter-industry wage differentials and employer size wage effects.;The second essay, "Taxpayer Utility and the Efficient Audit Rate: IRS Policy When Voters Dislike Being Audited," extends the model of efficient monitoring to the problem of taxpayer compliance when taxpayers are of two types, either honest or optimizing (tax cheats). It explores the determination of the optimal audit rate when the IRS has difficulty distinguishing taxpayer type. The model identifies the trade-off between the utility of additional government spending and the disutility of additional auditing and taxation. |